Comments

Hi Milford, First, you write: "When you read the systematic review by Mare Duvendack et al you will see it contains lots of references to impact evaluations that were undertaken using faulty methodologies, weak designs, little data, etc, etc, yet strangely ALL came out massively in favour of there being a positive impact from microfinance, with the main MFIs being most enthusiastic of all. That counts as very serious bias." I have now quickly read the systematic review. Throughout the term 'bias' is only used in a technical sense (typically selection bias), and while it is true that the authors note the poor quality of the evidence they reviewed, they themselves do not conclude that this is a product of a conspiracy to foist a particular view of economic development on the poor. Indeed (unless I missed it, in which case please point me to actual pages) while they do go so far as to actively caution against more microfinance on the basis of such a flimsy evidence base, they do not offer a systematic explanation of the poor quality of the work reviewed (I'm happy to offer some suggestions though: vested interests and ideology would be on the list as a possibility, but so too would be publication bias and the difficulties of conducting gold standard research in developing countries). Note too that you are incorrect to claim that all the reviewed studies in the systematic found strong positive impacts from micro-finance -- some did not. Second, you write: "You then quite ludicrously compare my brief blog comments, written to relax a little during a break from my regular academic and consulting work, with ‘the papers (not to mention Roodman’s book) (that) are careful and empirical and provide the reader with evidence and the tools for assessing whether their evidence is convincing’." I appreciate that blog comments are something that are, by necessity, typed in a hurry. But you have made very serious allegations about David Roodman and Dean Karlan, on the basis of no evidence whatsoever. If the evidence in question is in your papers or book, why can't you provide it in your blog comments here? If you don't have the evidence, then please cease making personal attacks in which you speculate on other researchers' motivations. I don't have a problem with more general debates about the intersection of research, knowledge, and power, but personal allegations need a basis in reality. Even more general blog comments come across as more convincing when provided coupled with evidence. I appreciate your attempt at providing the Duvendack systematic review, and your own book review in this spirit (and encourage readers to have a look), but in the case of the systematic review, as noted above, it still doesn't provide a basis for your boldest claims. Third on neo-liberalism, My complaint about the use of this term is threefold: 1. For many on the left, it often seems to serve as a magical form of punctuation that can be placed in a sentence to serve as a stand in for an actual argument. i.e. "X's argument is typical neo-liberalism". (maybe, but so what) 2. The term is also used as a catch-all for "bad stuff". Which, once again, doesn't strike me as helpful. To take the specific case of micro-finance. It isn't actually a free market approach to development (a true free market economist would believe that if the loans had use the money would have already been provided by efficient financial markets and taken up by those who needed it). And a sensible free market economist could make a good argument against micro-finance along the lines of: "when poor government policies are the main impediment to market led growth in a country it makes no sense to provide the poor with credit (and therefore debt)". 3. Even when neo-liberalism is restricted to something like a Washington Consensus suite of policies, I still don't think it can really be argued that this particular school of thought has failed per se. Some aspects (financial market deregulation, austerity during certain types of financial crises) surely have. But in some areas the results are more ambiguous (the outputs of privatisation vary from country to country and industry to industry). And in some areas the results seem net positive (an element of macro-economic stability). However, perhaps my complaints about the use of the term are unfairly levelled at you (although you're clearly a fan of the word). In which case apologies. Also, apologies in advance -- I have a hectic week and won't have much time to write blog comments. Thank you for the offer of sharing a pdf copy of your book. Terence

Bob: Thanks for this reminder. It's true that the IFC model probably cannot be 'reformed' sufficiently to meet the needs of the Pacific island countries without creating inefficiencies, and therefore that there is a role for a DFI operating in the sub-IFC range there. The government has so far announced two initiatives intended in different ways to increase private sector investment in the region: the SME-oriented <a href="http://adb.org/projects/details?page=details&proj_id=47373-001" rel="nofollow">Pacific Business Investment Facility</a> and, just recently, <a href="https://innovationxchange.dfat.gov.au/project/seed-pacific" rel="nofollow">SEED Pacific</a>, which aims to encourage larger businesses to invest (more on the latter next Monday, on the blog). Both these initiatives might well prove to be useful, but they are wholly or heavily reliant on grant funding and, related to that, quite limited in scale. For readers: Like Newton and Leibniz, Bob McMullan and Jim Adams hit upon the same idea independently. The former's proposal was made in a Devpolicy blog, <a href="https://devpolicy.org/development-finance-company-for-australia-and-new-zealand-20140117/" rel="nofollow">here</a>. The latter's was part of his 2013 Harold Mitchell lecture, <a href="http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2357288" rel="nofollow">here</a>.

Robin, You will recall that Jim Adams and I both advcoated a joint Australia/New Zealand DFI to focus on the Pacific. Therefore I welcome your consideration of the options. The problem with focussing all our efforts on the IFC is that they, quite understandably, have a broader focus and need a larger scale of investment than the Pacific Island states can sometimes generate although they have done some very good work in the Pacific. If the ADB had an effective private sector arm that might be an option but I believe only Australia nad New Zealand are likely to give the sustained attention to the Pacific which a successful DFI would require. The Austrians just set up a DFI funded by their EFIC equivalent but operating separately. This might be an option in difficult budgetary times although it is not ideal.

Thanks Paul. While ensuring the development quality of investments would be central, and was not always well done under DIFF, A DFI of the kind described above would not be the kind of animal that DIFF was. DIFF gave EFIC access to aid funds in order to soften the terms of its loans to developing countries, with those loans extended to governments and tied to the procurement of Australian goods and services. A DFI, by contrast, normally deals with the private sector rather than governments and finances or insures investments rather than just procurements (hence the fact that the US has both the Ex-Im Bank and OPIC). The question of tying, while a live one, tends to be rather moot in practice, as it is in the case of funding for donor country NGOs and at least some scholarships for tertiary study in donor countries. From a development perspective, it would certainly be better if donors deferred to IFC in this area, with appropriate policy and procedural reforms and oversight. However, the fact that we see a plethora of bilateral DFIs suggests that the market development perspective also looms large. There is a reasonable argument that putting ODA in the picture could leverage significant development impacts from the operation of such DFIs, but whether it would actually do so depends on how things are arranged at the outset.

Terence When you read the systematic review by Mare Duvendack et al you will see it contains lots of references to impact evaluations that were undertaken using faulty methodologies, weak designs, little data, etc, etc, yet strangely ALL came out massively in favour of there being a positive impact from microfinance, with the main MFIs being most enthusiastic of all. That counts as very serious bias. Indeed, as the authors contend, it was because of such bias that the neoliberal-oriented international development industry managed to raise up the microfinance model into a development policy juggernaut; that is, on the basis of an entirely missing evidence base an entirely new intervention became possibly the most important development policy intervention of all. The authors conclude that the only thing that can explain such an amazing story is the politics, not the economics. As many in the development agencies (the World Bank refuses to use the phrase, period), you appear to baulk at the use of the term ‘neoliberal’. Why? It is a standard academic descriptor of a particular market-driven worldview/project and it usefully helps to differentiate development policies based on other worldviews/projects (social democracy, democratic socialism, nationalism, etc) from those based on neoliberal principles and imperatives, as policies are in the World Bank, IMF, IAB, ADB, WTO, EuropeAID, USAID, DFID, etc. You then quite ludicrously compare my brief blog comments, written to relax a little during a break from my regular academic and consulting work, with ‘the papers (not to mention Roodman’s book) (that) are careful and empirical and provide the reader with evidence and the tools for assessing whether their evidence is convincing’. Maybe read some of my books and major papers first before making such a statement? You can find some of my main papers here: http://ssrn.com/author=2022134 and I’ll send the pdf of my ‘Why doesn’t microfinance work?’ book to you if you are genuinely interested. As to the evidence about Roodman’s book and its willingness to please the private sector funding bodies that supported and promoted his work, let’s get real here. It is beginning to happen all the time in academic circles, and, in fact, 'buying' academic support has always happened, especially in the US academic economics community: http://www.amazon.com/Economists-Powerful-Convenient-Distorted-Economics/dp/0857284592 so this is nothing new. It is therefore one obvious explanation – I would argue the main one - that can logically account for Roodman's strange book, a book that, as you say, first makes the important claim that ‘microfinance does not work to reduce poverty’ but then goes on to make a series of spurious arguments to the effect that we should nevertheless persist with its deployment! Roodman is a very smart cookie indeed and when he makes illogical twists and turns like that, which even such as Dean Karlan have called him out on (see my review of Roodman's book in RRPE), there must be some reason to account for it. Finally, to the issue of ad hominem attacks, it perhaps feels like that to you, but it is not. I am merely challenging the ideas, assumptions, outputs and methodologies of many of those working in the microfinance industry. However, because such factors are also wrapped up in other more messy things like the funding of research, the politics of development, power issues, and so on, it is not easy to separate the two sides. But your overarching idea here, that we should not challenge the motives and integrity of academic researchers, is simply wrong: finding out why someone writes what they write is often as important in explaining something as what they write. Unfortunately, as big business and Wall Street further and further insinuates itself into the academic community, especially in the USA and UK, we are likely going to see even more such situations. Milford

I am sure that the visitors to Tonle Sap lake believe what they think they saw. However a couple of days in Siem Reap does not give a true picture. 4 years ago on a holiday,I visited a small, very poor orphanage. Today I sponsor the same orphanage except we now care for 40 children, 6 paid staff and rent a comfortable building... If you could see some before and after photos you too would see how far these beautiful children have come.I am very grateful to any tourists who donate rice (we eat at least 200kgs a week) and 35 children go to school 6 days a week...See Heart and Love center Siem Reap....Helen Mclean, Victoria Australia

Hi Robin Possibly a very interesting development. Three quick comments. First, Australia’s experience with a mixed ODA/EFIC loans arrangement was of course the Development Import Finance Facility. In addition to remembering the lessons from previous evaluations of that scheme, it could be useful to have a new evaluation on the sustainability and development impacts of supported activities – such as the $60 million White Industries Piparwar coal mine in India and the $140 million Transfield Steel Bridges programs in Indonesia. Second, before creating any new institution, it is important to consider if a strengthened partnership with the World Bank’s International Financial Corporation or some similar entity may produce better outcomes. Always nice to have a new institution – but they are expensive and the skill sets required are demanding. Finally, just to note that EFIC has used its national interest provisions to assist with elements of development beyond simply export promotion. Its USD350 million of financing support to the PNG LNG project is a notable example. Its USD90 million risk sharing agreement with the Asian Development Bank has development linkages. Of course, these activities also have benefits for Australian trade facilitation, and balancing competing objectives will be a major challenge, especially with the commendable policy of untied aid. Paul

Hi Milford, Thank you for your continued engagement. To start on a positive note: thank you for the link to the DFID systematic review - I am looking forwards to reading that. In the meantime though, can you please provide me with particular page numbers where the authors show evidence of 'deliberate bias' rather than methodological flaws? I read no suggestion of this in the executive summary. Beyond the DFID review (which does look excellent, thanks again) I'm open to persuasion, and certainly willing to accept that ideology and vested interests shape development practice and the world of development thought far more than they should. (In the particular case of ideology, this is not, needless to say, a problem limited to the right, hence my jab at your persistent use of the term neo-liberal as if it were some form of data point). However, the reason why I found the papers that I linked to in this blog post useful, and why I do not find your comments persuasive, is that the papers (not to mention Roodman's book) are careful and empirical and provide the reader with evidence and the tools for assessing whether their evidence is convincing. You on the other hand (in your blog comments at least - the book review you linked to last night was more interesting) offer nothing more than speculation based on evidence which isn't transparent and which cannot be assessed by a neutral reader i.e. "Many people I have talked to, including one former CGD employee, broadly agree with me that his ‘review’ of my book was nothing of the sort: it was a carefully calculated smear attempt designed to kill my book." Similarly, when you write -- "As to the issue of who funded Roodman’s work, I think it is quite naïve to automatically assume, as you seem to do, that all researchers/evaluators produce independent work totally uninfluenced by those who fund their work." -- the big problem is that, while you are correct to suggest that funding can shape findings, you provide no evidence whatsoever that this was the case with regards to Roodman's work (work which, you will recall, was ultimately negative with regards to its assessment of microfinance). You have made very serious allegation on the basis no evidence whatsoever. I suspect that if we were to have a chat about development over a beer at a pub one day, we would find a lot to agree about: I do not think unimpeded markets are particularly wonderful tools for fostering economic development (let alone human well-being); I suspect industrial policy is a necessary but not sufficient pathway to development; I'm a micro-finance sceptic; and I think aid quite often goes awry as a response to the ideology of the day, not to mention the self interest of donors. But in this blog comments thread at least, you aren't persuading me at all -- mostly because you are engaging foremost in unsubstantiated ad hominem attacks against people you don't like. And because your critique comes across (perhaps unintentionally) as rich in its own particular dogma. Thanks for engaging though. And hopefully one day we might be able to have a more amiable chat away from the hurried world of blog comments. Terence

Hi Bill, Thank you for your comment. You make a very important point, I think -- that there is considerable difference between purely commercial micro-lenders, micro-lenders who offer subsidised/discounted loans (and whose motives aren't predatory), and other tools such as savings groups. To be clear, the studies I've linked to are not focused on savings groups (or at least primarily focused on undertakings that are based on savings groups). And as such this post shouldn't be taken to speak in anyway to the efficacy or not of such groups. Your point about subsidised microfinance as a means of undermining predatory lending very interesting point (thank you!). I guess, for me, something akin to this was one of the more interesting elements of the Rajbanshi et al paper was the point they made about a range of micro-credit operations usually being present in any one location, and how this makes it hard to gauge the impact of subsidised (or at least non-predatory) micro-credit efforts (the type that usually get evaluated). Thanks again. Terence

interesting to note that BPNG holdings of Tbills has gone from PGK200m in Jun-14 to a whopping PGK1.5bn by end Dec-14. I wonder what it would be by now?

Terence An interesting if a somewhat unfair response ('horrible neoliberal types'...). Anyway, a critique often involves addressing the work of high-profile people, but if I critique I always present the evidence underpinning my critique. Roodman’s critique of microfinance succeeded in one respect, but failed at multiple levels, as I wrote in my review of his book. Many people I have talked to, including one former CGD employee, broadly agree with me that his ‘review’ of my book was nothing of the sort: it was a carefully calculated smear attempt designed to kill my book. His debating points were all about typos, grammar, a missing reference, etc, while he completely refused to address the main arguments presented in my book, a book that has since been proven correct in almost everything it argued, as even very high-profile practitioners now <a href="http://zed-books.blogspot.com/2013/06/why-doesnt-microfinance-work_6.html" rel="nofollow">openly accept</a>. Perhaps, as several people suggested to me, Roodman was miffed that I had taken the wind out of his sails by declaring well before his own book came out that microfinance was a failure? I don’t know. But the individual clearly being smeared surely has a right to respond, don’t you agree? As to the issue of who funded Roodman’s work, I think it is quite naïve to automatically assume, as you seem to do, that all researchers/evaluators produce independent work totally uninfluenced by those who fund their work. This is wrong. You must have seen the Oscar-winning documentary ‘Inside Job’ for example. I also know full-well from my own work this last 20 years evaluating SME and local development programs for many of the big development agencies, that evaluations and evidence are distorted all the time. So my point that his ultimately very strange and contradictory conclusions were suspiciously almost exactly what his funding body would have wanted is a valid point to explore in order to explain what he did there. A little later, moreover, many other long-standing evaluations and evaluation evidence was <a href="https://www.givedirectly.org/pdf/DFID_microfinance_evidence_review.pdf" rel="nofollow">exposed</a> by Maren Duvendack et al in their systematic review of the evidence on microfinance, which found a lot of deliberate bias by MFIs and their contracted academic researchers all in favour of coming to a positive conclusion when the evidence for this was simply not there. As to my considered view that the impact evaluation by such as Dean Karlan and others are suspiciously biased on the upside, I already presented my case in the blog site I quoted in the earlier post. I think it is a strong case, and others actually working in microfinance have <a href="http://blog.microfinancetransparency.com/disguised-mediocrity-the-quest-for-positive-impact-results-at-compartamos/" rel="nofollow">gone even further</a> than me on this issue, such as Hugh Sinclair. In fact, there is actually a long history of individuals and institutions in the development and academic community deliberately misrepresenting the evidence, including their own research, in favour of coming to politically-acceptable outcomes or not to upset a funding body or to ingratiate themselves in with a corporation. One would have to say here, in fact, that the World Bank are particularly guilty on this point, of often grossly misrepresenting and distorting the evidence in favour of promoting a particular political viewpoint and agenda, as Fine et al showed in a <a href="http://www.amazon.co.uk/The-Political-Economy-Development-Neoliberalism/dp/0745331033" rel="nofollow">brilliant book</a> from a few years ago. So I am not saying anything at all radical here. Finally, to your last point. I am not saying that the political ideology point explains absolutely everything about why microfinance became so popular in spite of the total lack of evidence that it actually works, but it certainly explains almost everything (to misuse a phrase used by Paul Krugman). Anyway, I found the Rabanashi paper interesting, but ultimately weak and rather strange for addressing an important issue without mentioning the elephant in the room issue, the political-ideology motive for distorting the evidence to the upside. Milford

Hi Milford, Thank you for your comment. However, I think you do yourself no favours in the way you attempt to critique David Roodman and Dean Karlan. Both are careful empiricists and do not deserve the aspersions cast their way. In Roodman's case the fact that he qualified his critique seems very sensible to me. In the complex world of development it is very reasonable to assume that effects may be heterogeneous across locations, across populations within locations, and across programme designs. What is more, you accuse Roodman of trying to smear you (readers can make up their own minds: the relevant Roodman post is <a href="http://www.cgdev.org/blog/why-doesn%E2%80%99t-milford-bateman%E2%80%99s-book-work" rel="nofollow">here</a>) before alleging that Roodman qualifies his conclusions regarding microfinance because "[t]his part of his argument is probably what he was paid by Mastercard Foundation and CGAP". Do you have any evidence whatsoever for this allegation? Likewise do you have any evidence that Dean Karlan and other authors have reached particular conclusions because they have been motivated by funding considerations? It is one thing to disagree with others findings' but it is something else altogether to make accusations with regards to their motives in the absence of actual evidence. Finally, with regards to the only actual substantive point you make in your comment. For what it's worth I think it is possible that ideology does drive some people's continued enthusiasm for microfinance in the face of evidence that it is, on average, not particularly effective. But I doubt this is the only reason (not everyone who has a different take on development from you is necessarily one of these horrible neo-liberal types that you write so heatedly about) and I would urge you to pay attention to the Rabanashi et al paper that I linked to: it provides a pretty good explanation (and some evidence) of how people can overstate the benefits of micro-finance in good faith as a result of simple before and after comparisons. Kind regards Terence